INTRODUCTION
Mechanical thrombectomy achieves high initial reperfusion in acute large vessel occlusion (LVO), but intracranial atherosclerotic disease–related LVO (ICAD-LVO) is more prone to intraprocedural and early reocclusion than cardioembolic LVO, often necessitating adjunctive percutaneous transluminal angioplasty (PTA) or intracranial stenting with antiplatelet therapy [1-5]. Although PTA and stent placement are widely used as rescue strategies, they can carry periprocedural risks, including perforator infarction and hemorrhagic complications, and restenosis may occur after stenting. Together, these concerns highlight the need for safer, reproducible approaches for ICAD-LVO.
Temporary stent retriever deployment (TSRD) has emerged as an alternative strategy that enables temporary mechanical dilation of the stenotic segment using a deployed SR, while potentially limiting vessel injury and allowing time for antiplatelet therapy to take effect during deployment [6-11]. However, operators currently lack established criteria to decide between retrieving the device versus resheathing it. Intraprocedural decision-making relies heavily on individual operator experience, which may lead to inconsistent outcomes: premature retrieval may leave unstable lesions prone to reocclusion, while indiscriminate rescue stenting increases hemorrhagic risks. During standard thrombectomy in ICAD-LVO, operators encounter specific angiographic phenomena, such as in-stent clot formation or distal flow alterations. While these are well-recognized intraprocedural findings, a standardized framework utilizing them to guide the critical decision between stent retrieval and resheathing during TSRD remains unestablished. Therefore, we systematically evaluated real-time angiographic signs during TSRD, integrating them as objective indicators. Ultimately, we propose a stepwise procedural guide to standardize intraprocedural decision-making for ICAD-LVO.
MATERIALS AND METHODS
Patient Selection
Consecutive patients treated between April 2023 and September 2025 at the 2 participating institutions were retrospectively identified if they underwent TSRD during endovascular therapy (EVT) for ICAD-LVO. ICAD-LVO was diagnosed intraprocedurally and defined as residual fixed stenosis greater than 50% at the culprit segment after reperfusion or early reocclusion with an underlying focal stenosis at the same segment, based on the treating operator’s angiographic assessment [3,4]. Baseline stroke severity was assessed using the National Institutes of Health Stroke Scale (NIHSS), and early ischemic changes were evaluated using the Alberta Stroke Program Early CT Score (ASPECTS). Time metrics, including door-to-puncture time, were also recorded to contextualize the interventional workflow and establish the acuity of the treated cohort.
Procedures
Experienced neurointerventionists performed all procedures with patients under local anesthesia with conscious sedation. An 8-French (Fr) long sheath was placed in the right femoral artery, and an 8-Fr balloon guiding catheter was advanced to the target cervical artery according to the occlusion site. In cases where clinical and imaging features were suggestive of an underlying ICAD, systemic heparinization was administered to target an activated clotting time of ≥250 seconds during the procedure. Periprocedural antiplatelet management followed institutional protocols for suspected ICAD-related lesions and included an off-label loading dose of dual antiplatelet therapy (aspirin 200 mg and prasugrel 20 mg) or single antiplatelet therapy (prasugrel 20 mg). Regarding adjunctive medical therapy during the procedure, our standard approach included the uniform administration of intravenous antiplatelet therapy (ozagrel sodium). Intra-arterial thrombolysis was not administered in any of these cases. Antiplatelet agents were initiated when ICAD was confirmed on digital subtraction angiography (DSA) or when preprocedural imaging and clinical findings strongly suggested ICAD. After EVT, continuous intravenous argatroban was administered for 5 days, and aspirin (100 mg) and prasugrel (3.75 mg) were administered for 3 weeks, followed by single antiplatelet therapy.
A microcatheter was navigated distal to the occlusion, and a commercially available SR (Trevo; Stryker or Solitaire; Medtronic) was deployed across the stenotic segment. No detach-capable SRs were utilized in this cohort. TSRD was intended to be maintained for up to 30 minutes, with DSA performed every 10 minutes during deployment to evaluate antegrade flow. For this exploratory study, in-stent thrombotic sign (IST) and post-resheathing distal flow deterioration (PR-DFD) were prespecified as angiographic signs based on the presumed pathophysiology of ICAD-LVO during TSRD. IST was considered to reflect an active thrombotic milieu at the culprit plaque, potentially caused by the interaction between stent struts and a disrupted atherosclerotic surface. It was defined as either intra-stent occlusion or an intra-stent angiographic filling defect on the scheduled DSA runs. In contrast, PR-DFD was hypothesized to reflect the loss of temporary radial support after resheathing, resulting in elastic recoil, impaired antegrade flow, distal hypoperfusion, or microthrombotic flow disturbance. PR-DFD was defined as a noticeable delay or reduction in blood flow distal to the lesion observed after stent resheathing (or prior to procedure completion), compared to the optimal distal flow achieved immediately following initial maximal stent deployment. These signs were therefore interpreted not as validated biomarkers, but as real-time angiographic cues intended to support intraprocedural risk assessment.
Procedural strategies were tailored to these angiographic findings at the operator’s discretion. Typically, when an IST was identified, the strategy was converted to stent retrieval for thrombectomy. In the absence of such findings, the SR was resheathed while maintaining distal microwire access, and the microcatheter was withdrawn proximally to reassess the presence of PR-DFD. The presence of these angiographic signs was evaluated intraprocedurally by the primary operator and a second neurointerventionist, with discrepancies resolved by consensus.
Outcome Measures
The primary outcome was the acute procedural success of the temporary stenting strategy, defined as the successful securing of target lesion patency during the endovascular procedure and the stable maintenance of that patency thereafter. Other angiographic and clinical metrics were evaluated as secondary outcomes or safety endpoints; these included successful final reperfusion, defined as a modified Thrombolysis in Cerebral Infarction (mTICI) score of 2b–3 at the end of the procedure, early neurological deterioration (END), defined as an increase in NIHSS score of ≥4 within 7 days after EVT [12], and reocclusion on follow-up vascular imaging within 7 days. To ensure consistency between procedural classification and clinical results, it is important to note that intraprocedural angiographic outcomes were evaluated independently from clinical outcomes. Specifically, cases with acute reocclusion requiring rescue balloon angioplasty were classified strictly under the angiographic category of PR-DFD. Conversely, END was evaluated as a strictly distinct clinical outcome. Hemorrhagic complications were classified according to the Heidelberg bleeding classification [13]. Symptomatic intracranial hemorrhage (sICH) was defined as any intracranial hemorrhage accompanied by an NIHSS worsening of at least 4 within 24 hours. Procedural time was measured from groin puncture to final reperfusion. Total endovascular maneuvers were defined as the sum of stent retrieval passes, TSRD deployments, and rescue balloon angioplasty attempts. Ninety-day outcome was assessed using the modified Rankin Scale (mRS), with favorable outcome defined as mRS 0–2.
RESULTS
Patient Characteristics
A total of 8 patients (3 men; median age, 72 years) were included in this study. The occluded vessels were the M1 segment of the middle cerebral artery in 4 patients, the M2 segment in 2, the internal carotid artery in 1, and the vertebral artery in 1. The median baseline NIHSS score was 11, and the median baseline ASPECTS was 9. The median door-to-puncture time was 97 minutes. Regarding the diagnostic pathways, 6 patients (75%) had preprocedural features suggestive of having underlying ICAD prior to the procedure. The specific preprocedural features leading to this determination were the complete absence of atrial fibrillation combined with a characteristic internal border-zone infarction on diffusion-weighted image. All 6 of these patients received antiplatelet therapy before the procedure. In contrast, the remaining 2 patients (25%) had no suggestive preprocedural features. However, all 8 patients—regardless of their preprocedural suspicion—ultimately met the identical intraprocedural angiographic criteria defined in the MATERIALS AND METHODS section, which definitively confirmed the diagnosis of ICAD-LVO. The baseline characteristics of the 8 patients are summarized in Table 1.
Procedural Findings and Outcomes
The median number of TSRD attempts was 1 (range, 1–2), and the median duration of TSRD was 20 (range, 15–30) minutes. The median total procedure time was 65 (range, 35–157) minutes. The IST was observed in 3 patients, all of whom subsequently underwent stent retrieval for thrombectomy and achieved sustained reperfusion. In the remaining 5 patients without IST, the SR was resheathed. Among these 5 patients, PR-DFD occurred in 3:1 angiographic reocclusion requiring rescue PTA and 2 persistent distal flow delays without reocclusion. During the intraprocedural observation period, there was no instance of delayed IST or PRDFD. Regarding the primary outcome, all 8 patients (100%) achieved successful securing of target lesion patency during the procedure, which was stably maintained without acute intraprocedural reocclusion except for the 1 patient who required rescue PTA to restore patency. Secondarily, all 8 patients ultimately achieved successful final reperfusion (mTICI 2b–3) at the end of the procedure without sICH. No postoperative reocclusion was observed. END occurred in 2 patients with PR-DFD without reocclusion. At 90 days, 5 of 8 patients achieved favorable outcomes (mRS score, 0–2).
Illustrative Cases
Representative cases demonstrating the angiographic signs and our stepwise approach are provided. A case of IST managed with stent retrieval is illustrated in Fig. 1. Cases managed with resheathing, including those with and without PR-DFD, are detailed in Supplementary Material 1, with corresponding imaging provided in Supplementary Figs. 1 and 2. Furthermore, the real-time angiographic progression of PRDFD is demonstrated in Supplementary Video 1.
Narrative Literature Review
A summary of previously reported cases utilizing temporary stent retriever angioplasty for ICAD-LVO is provided in Supplementary Table 1.
DISCUSSION
In this small 2-center retrospective series, we describe 2 angiographic signs observed during TSRD for ICAD-LVO and propose a practical, hypothesis-generating framework for intraprocedural assessment (Fig. 2). The principal finding is not that IST or PR-DFD alone should mandate a specific rescue strategy, but rather that these signs may help operators distinguish 2 different procedural problems: thrombotic instability during stent dwell time and hemodynamic instability after the loss of stent-mediated radial support. This distinction may be clinically useful because each scenario may require a different therapeutic response.
Pathophysiologically, IST likely represents in situ platelet activation triggered by the stent struts interacting with a vulnerable, ruptured plaque [14]. Resheathing the device when IST is present risks leaving unstable residual thrombus, potentially causing early reocclusion or distal embolization. IST should not be interpreted as an automatic indication for repeated stent retrieval. In our series, a single retrieval maneuver after IST was technically successful, but repeated retrieval in ICAD-LVO may aggravate endothelial injury, plaque disruption, and recurrent thrombosis [5,15]. Therefore, the presence of IST should be viewed as a warning sign of active thrombus formation rather than as a stand-alone treatment rule. When IST is observed during our routine 10-minute angiographic checks, a limited initial retrieval may be reasonable if intraluminal thrombus is suspected. However, recurrent IST or persistent flow compromise should prompt consideration of alternative strategies, including intensified antiplatelet therapy, balloon angioplasty, or rescue stenting, depending on hemorrhagic risk, infarct burden, lesion morphology, and operator judgment.
Conversely, in patients without IST, continued TSRD with antiplatelet loading facilitates stable reperfusion. Resheathing rather than retrieving the device minimizes additional endothelial injury [2,4,5], and all such patients in our cohort achieved successful reperfusion. However, PR-DFD—which may reflect elastic recoil, microthrombosis, or microcirculatory dysfunction—can still occur. The emergence of PR-DFD is strongly associated with early reocclusion or neurological worsening [16]. PR-DFD differs from conventional recognition of reocclusion tendency in 2 respects. First, it is defined by a temporal comparison between the best antegrade flow during maximal stent deployment and the flow immediately after resheathing, rather than by a static assessment of residual stenosis or delayed complete reocclusion. Second, it is assessed while distal microguidewire access is intentionally maintained, allowing immediate escalation to rescue angioplasty or other adjunctive therapy before complete reocclusion occurs. Thus, PR-DFD is not intended to replace established criteria for rescue therapy in ICAD-LVO; rather, it may provide an earlier angiographic warning of insufficient distal perfusion after removal of temporary radial support.
In our series, PR-DFD occurred in 3 patients: 1 required rescue PTA, while 2 received no additional intervention. Although the latter 2 avoided early angiographic reocclusion, they suffered postoperative infarct expansion, END, and poor outcomes. Retrospectively, their distinct PR-DFD indicated severe hemodynamic compromise insufficient to salvage the penumbra. Inadequate perfusion after stent removal may be associated with long-term functional failure. Therefore, observing PR-DFD may warrant lowering the threshold for adjunctive treatments, such as rescue angioplasty, to secure sufficient distal perfusion and halt infarct growth, even if the vessel appears angiographically patent.
This study has several limitations. First, it is a small retrospective case series (n=8) without core-laboratory adjudication, making the findings hypothesis-generating. Second, the Japanese antiplatelet regimen may limit generalizability. Third, short TSRD duration and 10-minute imaging intervals may miss evolving changes. Finally, IST may be difficult to detect; therefore, residual stenosis, stent expansion, and intra-stent findings should be evaluated together [17,18].
CONCLUSION
Our findings suggest that these well-recognized angiographic signs, specifically IST and PR-DFD, may provide practical cues for estimating thrombogenic burden and hemodynamic instability during TSRD. Although preliminary, structuring these observations into a stepwise guide offers a conceptual foundation to transition from individual operator experience toward more standardized intraprocedural assessment.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
Print
